Isaac sim 2020 remote server deployment problem

Omniverse Isaac Sim
1

Hi,

I’m trying to deploy Isaac sim 2020.1 EA on a remote Aliyun server (located In China). The server configurations are shown as follows with command ‘nvidia-smi’:
±----------------------------------------------------------------------------+
| NVIDIA-SMI 440.100 Driver Version: 440.100 CUDA Version: 10.2 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla V100-SXM2… Off | 00000000:00:08.0 Off | 0 |
| N/A 35C P0 54W / 300W | 0MiB / 16160MiB | 1% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| No running processes found |
±----------------------------------------------------------------------------+

I tried to deploy isaac sim using container. However, the image pulling speed is extremely slow. I gave this up after 2 day’s pulling. So, my first problem is 1. Is there any solution that can speed up the pulling process?

Besides, I tried to run natively on the remote server using the command ’ ./isaac-sim-2020.1.1920-4b1097bc-linux-x86_64-release/_build/linux-x86_64/release/isaac-sim-headless.sh --allow-root’. However, the error occurs as follows:

image
image1420×668 129 KB

So, 2. Is there any solution to address this error?

Many thanks!
Qin

2

Hello, and welcome to the IsaacSim community. I am not an expert in the problem you describe, but I suspect the Python environment is not right. The difficulty in configuring the environment is a major reason why we offer a containerized solution. It can be very difficult to debug this sort of problem. I will consult with engineers who might have insight into your situation.

3

Hi Qin,

Issue #1
If it is possible, could you try running it on a server located in different location, maybe the US? You could also try using a VPN connection to the US?

Issue #2
Could you please post a log with a -v flag to the command? This could produce more verbose logging that help us narrow down the problem.

4

Thanks for your reply! I’ll try to switch to the containerized solution.

5

Hi, thanks for your reply.

I have tried to use a VPN, unfortunately had no obvious effect.

The log with a -v flag is shown as follows:

image
image1659×463 84.9 KB

However note that the error is not the same as the previous post. I cannot reproduce the previous error probably because I have installed all the dependencies that Isaac SDK requires.

6

That error indicates that the GPU is not supported. We currently recommend running on an RTX series card such as a 2060 and above or a Tesla T4.

For running on the cloud, currently we support Amazon EC2 with the Tesla T4 GPUs.

7

Understood. Thanks!